Estimating Pathway Inhibition

Question

[miquelduranfrigola]

Hi all,

imagine that I have a pathway P (e.g. from KEGG). The only information I have is the topology of P (nodes, edges, and direction of the edges).

Say that now I want to inhibit (modulate, if you wish, for a more general case) P. I achieve my inhibitions by formally removing edges from P. Intuitively, removing an edge that is upstream should cause a higher degree of inhibition, and possibly the same would happen for edges that depart from central nodes, and so on. However, if you take a KEGG pathway you will soon realize that the casuistics are huge, and non-trivial.

Are you guys aware of some method or approach to account for all this? As I said, I only have the pathway topology, and I want to estimate the level of pathway inhibition I would achieve by removing certain edges. I'm not an expert on the field: is this what you call "network dynamics"? Any hints will be more than welcome!

Thanks.

Answer 1

[Khader Shameer]

Signaling Pathway Impact Analysis (SPIA; see package here) could be an approach that could be similar to what you are looking for.

Another approach is to treat your pathway as a network and perform network level perturbation analysis. NEXCADE or your network library of choice with a bit of scripting and stats could do such analysis. See related manuscripts 1 and 2

[miquelduranfrigola]

Great! Thanks Khader.